What Is AI and What Can It Do in NPD for You and Your Business?

Dr. Robert Cooper

kHub Post Date: January 3, 2025

Read Time: 10 Minutes

AI and Product Innovation

What AI can do for you and your company in new product development (NPD)? That’s the big question today. Many senior management teams seem uncertain about the ultimate role of AI in their business, its potential impact, and whether they should move forward at full speed, or just put their toe in the AI water. Similarly, many people working within NPD—ordinary employees and line managers—are also somewhat confused: They wonder what the role of AI will be, its impact on their jobs, and in some cases, why the company has not moved faster to adopt some of the more obvious AI applications in product innovation.

AI is a Super Prediction Machine

Let’s start with what AI is, and what AI is capable of doing. Gemini, Google’s online generative AI model, defines AI like this: “AI is technology that allows machines to learn and perform tasks typically requiring human intelligence, like learning from data, solving problems, and making decisions. AI encompasses various techniques like machine learning, deep learning, and natural language processing, enabling machines to extract insights from vast datasets, automate repetitive tasks, make data-driven decisions; and predict future outcomes.”

Moving beyond this technical definition, a more pragmatic approach is offered by Agrawal, Gans, and Goldfarb in their best-selling book [[i]]: AI is a prediction machine. It excels at using data to make faster, more accurate, and cost-effective predictions than any other method. When your problem involves predicting outcomes, like market size or sales volume for a new product, AI reigns supreme.

Even for non-predictive problems, AI can cleverly translate them into prediction tasks and generate solutions. Imagine a product developer seeking a new design for a concept car. By leveraging an AI model, like Open AI’s DALLE, they can transform the design challenge into a prediction: The result? A potentially groundbreaking design drawing—a testament to AI’s creative capabilities. Using AI with appropriate prompts can lead to a surprising array of novel and valuable new product ideas, concepts, and designs. Given that definition of AI as a prediction machine, the possibilities for AI in helping you and your company are endless.

The Power of AI for NPD

NPD is where AI excels. Here, AI plays a dual role:

1.     the process role, and

2.     the product role.

In the process role, AI acts as a silent partner, assisting at every stage of product development. For example, while AI for autopilots and driver assist may be the most visible applications of AI in autos, AI is also being used to design concept cars, and then predict consumer reactions to the new designs at GM [[ii]]. From ideation to market research, concept testing, engineering design and even planning the launch, AI capabilities streamline and enhance the entire innovation process (see Figure 1). Imagine AI algorithms sifting through vast data sets to uncover customer needs or generate design prototypes based on user preferences and feedback. These applications of AI not only expedite the development timeline but also ensure that products are finely tuned to meet the evolving demands of the market.

Figure 1. Three AI-target areas where AI has a huge impact in the NPD process—

refer to Tables 1, 2, and 3 for details of these applications

The front-end of NPD in Figure 1: AI applications currently in operation in companies in the front-end of the NPD process are outlined in Table 1, and include:

o   AI for idea generation;

o   concept generating, design, and concept testing;

o   market analysis;

o   voice of customer (VoC) research;  

o   competitive analysis;

o   technical assessment (including IP and source-of-supply); and

o   AI for creating the business case.

Many of these tasks are where project teams are often deficient: Either the tasks are poorly executed or worse, not done at all, yet they make all the difference between NPD success and failure [[i]]. If your objective is to increase the success rate of your NP projects, then more reliance on AI to handle some of these time-consuming front-end tasks is an obvious next step.

The front-end AI applications shown in Table 1 are not an exhaustive list, and the illustrations described are just a small sample of the many AI application examples in industry[1]. We refer you to other articles for more details on these applications [[ii]].

Development and Testing Stages: Here, many AI applications are very high impact on both the product and the process. Typical application are noted in Table 2, and include:

o   AI for engineering design and optimization;

o   product simulation and testing;

o   molecular and material discovery;

o   Digital Twins [[iii]]; and

o   AI for field trials, user tests, and product monitoring.

AI accelerates the design and development phases in some firms by over 50% [[i]]. Since these Development

tasks are often the most-costly and time-consuming ones in the entire NP project, acceleration here has major

payoffs [[ii]].

Applications range from engineering design for pistons in diesel engines, turbine blades in jet engines, aircraft airframe design, and transmission optimization in power trains. For example, GE researchers, using ML, trained a surrogate model with 100 computational models to determine the optimum shape for the crown of a piston in a diesel engine. This rapid iterative development created an optimal piston design that delivered a 7% improvement in fuel efficiency with a reduction in soot emissions…. all in 15 minutes!

In the world of materials products, AI applications include the creation of new consumer cleaning products (discovery of superior enzymes), chemical discovery (eco-friendly pesticides), and drug discovery (dozens of drug projects are currently underway using AI).

These varied and proven AI applications make them strong candidates for AI. What’s more, these Development and Testing AI applications have the strongest impacts on NPD KPIs, as shown in a previous kHUB article [7].

The back-end of NPD: Often overlooked is the potential for AI at the back-end of the NP process, but here many superb applications exist, as noted in Table 3 and below:

o   AI for creating the launch strategy and plan;

o   helping with elements of the marketing mix: pricing, marketing communications, distribution, and salesforce;

o   post-launch monitoring of product performance; and

o   AI for monitoring and analyzing customer feedback.

One of the most critical AI applications is for monitoring the product once in use to ensure customer satisfaction using “data based digital twins”. For example, every Telsa car sold has onboard sensors that relay performance data back to its digital twin, so the company can keep track of how well the vehicle is performing. GE does the same for its new jet engines once in flight.

Additionally, Natural Language Processing (NLP) can process feedback from customers—on blogs, customer

reviews, and customer complaints. NPL summarizes and makes sense out of this valuable feedback, taking unstructured text and identifying themes, even creating charts and figures. This is a huge opportunity to generate ideas for that next great new product.

Many AI applications exist for manufacturing and supply chain management at the back-end of the NPD process, but are beyond the scope of this article—a Forbes’ article outlines these [[i]].

The Product Role for AI in NPD

In most industry sectors, AI can be readily embedded within the product itself to yield unique user benefits, thus gaining competitive advantage [6]. These industries are where products are more complex, or based on electrical or mechanical engineering. For example, medical scanners such as Siemens’ AI-Rad Companion, use AI to analyze medical images, helping radiologists interpret scans more accurately and quickly; and in today’s autos, advanced driver assistance systems (ADAS) help drivers avoid accidents, such as lane departure warning and automatic emergency braking. In agriculture, witness self-driving tractors such as John Deere’s Autonomous tractor and Monarch’s EV tractor.

Examples of clever AI-embedded products are in Table 4, where AI enhances the product to yield a compelling value proposition, one of the keys to NPD success [6]. Note that the examples cut across a wide variety of industry sectors, from apparel to agricultural equipment… AI is quite inclusive. We make the point that embedding AI into your product is an option in most industries; one must be creative.

In process industries, where the end product is a material, chemical, polymer, or food and beverage, generally AI plays a greater role in the process of conceiving and developing the product rather than in the end product itself. For example, in the paint industry, it is relatively easy to see ChatGPT being used by the project team to create new ideas for paint, or even using BASF’s Emollient Maestro to “discover” a novel molecule—an environmentally-friendly emollient to improve paint flow.

It’s a stretch, however, to imagine how AI might be embedded in housepaint to add functionality. Or is it? By adding a QR code or an NFC tag (Near Field Communication tag) to the paint can, the painter can be linked to online painting tips and advice or real-time troubleshooting; and retailers could more easily manage their inventory.

For software development, which is not the focus of this article, AI cannot only be embedded in the end product, such as in Siemens’ Xcelerator CAD software with AI, but AI can be the end product or service itself, such as IBM’s Chef Watson or Microsoft’s Azure.

Process Versus Product AI

Currently, process-related benefits from AI likely hold a slightly larger share of AI’s total benefit in NPD than product-related benefits. This is mainly due to lower implementation hurdles, and a faster pay-back due to efficiency gains, cost reductions, and risk mitigation, which offer more immediate and quantifiable benefits.

Additionally, AI tools for process optimization and automation are generally more mature and readily available than embedding complex AI capabilities in diverse new products.

In the mid-to-long term, the pendulum is expected to swing toward product-derived benefits. This shift is driven by rising expectations of customers, who will demand smarter, more personalized products as in Table 4, and also by advancements in technology that will make embedding AI in products more feasible and cost-effective.

Based on many different reports and articles, a rough estimate would be 60–40 in favor of process benefits currently, shifting towards 50-50 or even 40-60 for product benefits within the next 5–10 years. The numbers are a guide, but the optimal split in relative emphasis for your business between NP-process and product-embedded AI, like any portfolio decision in NPD, depends on your firm, your industry sector, and your context.

Embarking on the AI Journey for NPD

AI is not a replacement for human ingenuity; it’s a powerful tool that can augment human capabilities. By embracing AI responsibly and strategically, companies producing physical products can unlock a new era of innovation, efficiency, and customer satisfaction. The future of business is undoubtedly AI-powered, and those who adapt will thrive in this exciting new era.

The time to start the AI journey is now… not next week or next month. The adoption of AI in business and NPD began some years ago, but like all previous technology revolutions, it started slowly but now is coming on like a huge tsunami wave and picking up speed. Currently, the “early majority” is adopting AI for NPD. The adoption window will be narrow, estimated to be about 22 years based on an analysis of previous technology waves, with the peak hitting before the end of the decade [[i]].

Given the short window, there is no time to waste. There are risks to adopting AI, but the risk of inaction is even greater. The only way to avoid obsolescence is by embracing innovation.

ABOUT THE AUTHOR

Dr. Robert Cooper, Professor Emeritus, McMaster University, Canada ISBM Distinguished Research Fellow at Penn State University A world expert in the field of management of new-product development and product innovation, Dr. Cooper has written 10 books on the topic and more than 170 articles. Bob is the creator of the globally-employed Stage-Gate (trademarked) process used to drive new products to market; a Fellow of the Product Development & Management Association; ISBM Distinguished Research Fellow at Penn State University. He is a noted consultant and advisor to Fortune 500 firms, and also gives public and in-house seminars globally.

LinkedIn: Dr. Robert G. Cooper
Website: www.bobcooper.ca

References

What did you think of this post?

Start a conversation with your peers by posting to our kHUB Discussion board! Browse trending posts and reply to other thought leaders OR start your own discussion by clicking "Post New Message."

Start a Discussion

If you don't have an account with us, create a guest account or become a member today and receive exclusive access to all PDMA member benefits. Please note that both members and non-members are welcome to participate in the kHUB.

Related Content